Exploring the Dynamics of Resilience in an Elementary School

Resilience Guide: A Collection of Resources on Resilience in Children and Families

by Saundra Murray Nettles and Frances P. Robinson

The School and Its Community

The School

Stanton Elementary School is located in southeast Washington, DC, in a highly commercialized area bordered by three major thoroughfares. As of February 1998, there were 620 students enrolled. According to the school's Title I Local School Improvement Plan, Update 1997-98, about 42% of the students are enrolled in prekindergarten, kindergarten, or first grade. All of the students are African American, and the median household income for the school catchment area is $12,000. Approximately 98% of the students are on the free or reduced lunch program. The school has a principal and a vice-principal, 29 teachers, one full-time counselor, a librarian, one building resource teacher who serves as the change facilitator, nine educational aides, a Title I Parent Coordinator, a part-time psychologist, and 15 support staff (such as clerical, cafeteria, and maintenance staff).

Stanton School has a history of engagement in two concurrent efforts toward creating a caring, supportive environment with high expectations for student success. First, since 1995, the school has undertaken activities which are components of the Comer process (Comer, 1985). The school is still designated a Comer school, and many of the outcomes that the Comer process is intended to achieve overlap with outcomes observed in studies of resilient children (see for example, Winfield, 1991; Werner, 1990; Nettles & Pleck, 1994). Although there has not been a formal evaluation of the Comer process at Stanton, when the resilience framework was introduced in the Spring 1996 semester, major elements of the Comer process were in place, namely community partners and the site-based school improvement team consisting of parents, faculty, and a mental health worker.

Second, the school had been identified during the 1996-97 school year as a targeted assistance school (that is, one needing program improvement to increase student achievement). Title I funds provide programmatic activities to improve student learning. The District of Columbia Public Schools (equivalent to the state education agency, or SEA) requires that the school consult with parents and submit a Title I improvement plan. The plan which was approved for the 1996-97 school year outlined activities toward goals for increased basic and advanced reading and mathematics competence, improved skills in writing, problem solving, and higher order thinking, heightened parent and community involvement, and enhanced professional staff development to reflect emerging reform issues. With the introduction of district-wide emphasis on improved reading and mathematics performance, Stanton's 1997-98 plan identified literacy as the number one priority.

The Community

Although Stanton serves students that come from low-income families, the school is located in a community that has diverse economic circumstances. According to the 1990 Census of Population and Housing, in the school's zip code 23% of families had incomes above $50,000; 29% had incomes between $25,000 and $49,999; and 47% had incomes below $24,000. Thirty-one percent (31%) of families with children under 18 lived below the poverty line; 42% of female-headed families with children under 18 lived below the poverty line.

There is diversity in occupations and educational attainment. Twenty-three percent (23%) were in executive, administrative, and professional specialty occupations. A substantial minority (42%) were in the combined categories of administrative support positions and service occupations. Of persons in the labor force, about 10% were unemployed (nearly twice the national average in 1990). Of individuals 18 years and older, 35% had not graduated from high school; 33% were high school graduates; 21% had some college education; and 11% had college degrees or graduate and professional degrees.

A variety of housing types may be found in the neighborhood; 86% of the units are occupied. Residents have pressed for economic development, and in 1997, a shopping center was built near the school. The center includes a large food store and several shops, such as a shoe store. In addition, several fast food chains operate within the blocks surrounding the school. The school staff and families have expressed concern about the extent of violence, other crimes, and drugs near the school. In short, the school and its setting have sources of protection and sources of risk.

Applying the Resilience Approach

Stanton's application of the resilience construct integrates and extends the Comer process and the school improvement priorities of Title I/Chapter 1. As discussed below, Stanton is using three strategies to apply the resilience approach: (1) implementing activities to increase resilience; (2) assessing paths to student resilience; and (3) increasing faculty, staff, and parent awareness about resilience and related constructs. We discuss each of the strategies below.

Implementing Activities to Increase Resilience

Stanton is implementing two sets of activities to foster resilience in students: (1) increasing available resources; and (2) mobilizing protective processes in the environment. These are basic strategies used in many interventions, such as Head Start. Evidence of their effectiveness comes from research on the "ingredients" of resilience as identified by Masten (1994). Among these ingredients are the risks or adversities the individual faces and the individual and environmental characteristics that serve as protection against risk. The following describes the specific ways in which the two strategies are being implemented at Stanton.

Increasing resources available to Stanton students. Increasing the resources available to students at risk is a basic activity for many schools. At Stanton, the activity includes the assembly of resources from the school district as well as the development of community partnerships. Stanton's community partners provide materials, funds, programmatic activities, volunteers, and services on behalf of Stanton's students. For example, instruction in life skills is provided for fifth and sixth grade boys through the Preteen Pregnancy Prevention Program for Boys (sponsored by Concerned Black Men); a Rites of Passage program for 20 fifth graders is sponsored by the Family Medical and Counseling Center; a mentoring program for 35 girls in grades five and six is conducted by the community-based organization Naje/SAFE; and mentoring for sixth grade boys is provided by PROJECT 2000, Inc.

The resources are distributed according to needs identified in plans developed by the school staff and the school improvement team. During 1995-96, resources generated from community organizations were used primarily to renovate the physical plant. For example, the Peace Corps and Trinity United Methodist Church, among others, painted bathrooms, the health suite, and the boiler room hallway. In 1996-97, physical improvements continued with assistance from community partners; however, a sizeable share of community resources were targeted to either motivate students to excel in academic areas or to increase mastery of reading and mathematics. During 1996-97 and 1997-98, additional community partners were recruited to provide, at each grade level, one-to-one tutoring or other forms of student assistance to augment the reading and mathematics programs. For example, volunteers from The Pentagon ("Book Buddies") provide weekly, one-to-one reading experiences for first graders. One church partner conducts The Summer Intensive Reading Program for students in grades three through six, and volunteers from a second church partner provide tutoring and remediation in reading and other subjects during the school year. Incentives are offered by two local science and technical organizations, Diversified Engineers and SMART. The Social Security Administration, through its national YouthLink Program, provided 60 computers and software for access to the Internet. Community partners, such as The Pentagon Book Buddies, helped to wire the school. The resources and partners (shown in parentheses) are identified in Table 1.

Mobilizing protective processes. The second set of activities Stanton is using to foster resilience in students is the enhancement of protection in the environment. The activities, shown in Table 1, are grouped according to Benard's (1991) three categories of protective processes: high expectations, opportunities to participate, and caring and support.

Table 1
Activities to Mobilize Protective Processes in the Environment

PROTECTIVE PROCESS: ESTABLISHING HIGH EXPECTATIONS

GRADE

ACTIVITY

EXPECTED RESILIENCE OUTCOMES

K-1

1

2, 3

National Institute of Child Health (NICH)

Book Buddies (The Pentagon)

Sylvan Learning Systems, Inc. (Reading)

All reading activities: Reading motivation, greater proficiency in reading; movement of children from below basic reading levels to basic level and above

3-6

Reading Resource Teacher Summer Intensive Reading Program (Allen A.M.E. Church)

 

2-6

Tutoring and remediation (Trinity United Methodist Church; U.S. Navy Kids Program; Law firm of Pepper, Hamilton, & Scheetz)

 

5

1

Schoolwide

Incentives for Science Achievement

D.C. Reads (Tutors from American University)

Computers/Internet Access (Social Security Administration, NASA, Department of Transportation)

 

Schoolwide

Schoolwide (Staff)

Technology Program

Standards Specialist Coordinator

 
 

PROTECTIVE PROCESS: PROVIDING OPPORTUNITIES TO PARTICIPATE

GRADE

ACTIVITY

EXPECTED RESILIENCE OUTCOMES

Schoolwide

WSES (simulated radio broadcast)

Extracurricular activities (Drama, Double Dutch, Safety Patrol, Dance, Choir, Pep Squad, Substance Abuse Program, Student Council, Basketball, Red Cross)

Exposure to soccer, golf, and tennis as part of physical education program (PROJECT 2000, Inc., Concerned Black Men)

Trips to cultural events (Concerned Black Men, Washington Performing Arts Society, Kennedy Center)

Community service projects

Greater proficiency in reading and speaking

Social competence

Sense of responsibility to others

 

PROTECTIVE PROCESS: CARING AND SUPPORT

GRADE

ACTIVITY

EXPECTED RESILIENCE OUTCOMES

Schoolwide

Programmatic & financial support (Allen A.M.E. Outreach; Trinity United Methodist Church; PTA; Concerned Black Men; Law firm of Pepper, Hamilton, & Scheetz)

Incentives for achievement (SMART: Science Mathematics Aerospace Research Technology; Diversified Engineers)

Increased resources

Pre-K

Programmatic & financial support (Children's Educational Fund)

 

3

Programmatic & financial support (Council of Women's Ministries)

 

5-6 (girls)

6 (boys)

Mentors (Naje/SAFE)

Project 2000, Inc.

Conflict resolution/peer mediation (Law firm of Pepper, Hamilton, & Scheetz)

Drug/Alcohol Program (Allen A.M.E. Outreach)

Substance Abuse Prevention Program (SAPE, Title IV funding)

Social competence

4

Learning for Life (Boy Scouts of America)

 

5

Rites of Passage (Family Medical & Counseling Center)

 

4-6

Preteen Pregnancy Prevention Program for Boys (Concerned Black Men)

 

Schoolwide (Parents)

Parent Outreach Program

PTA, Parent Academy

Increased parental support for children's education

Schoolwide  (Staff)

Dissemination of Stanton Monthly Bulletin

Higher morale among staff

Because one of Stanton's priority goals is to foster competence in academic areas, especially reading, many activities which supplement existing instruction have been introduced to increase students' mastery experiences.

Expected student outcomes include enhanced self-efficacy and greater motivation for and proficiency in reading. Such activities, as shown in Table 1, are to mobilize the protective process of establishing high expectations. For example, the kindergarten and first grade implemented a reading program developed through the National Institute for Child Health and Human Development. The program emphasizes phonemic awareness, literature, and phonics. Volunteers (parents, grandparents, and community partners) participate in the program's reading activities. In grades three through six, the reading resource teacher conducts reading instruction in small groups of the lowest achievers. The Sylvan Reading Lab, a component of the Sylvan Learning Systems program, serves students in the primary grades. The Lab includes special equipment for drill and phonics and for increased reading comprehension, and instruction consists of the Sylvan curriculum, which focuses on basic reading skills.

The second protective process is providing opportunities to participate; the activities are shown in Table 1. One daily activity is Radio Station WSES (Wonderful Stanton Elementary School). In keeping with the school's focus on reading competence, this is a five-minute "broadcast" from the principal's office to all classrooms in the school. Students in Head Start through sixth grade rotate as announcers of schoolwide news, birthdays of students and staff, and reading of literary selections and thoughts or affirmations of the day. The theme song for the broadcast is Ray Charles' rendition of "America the Beautiful." Other opportunities for participation include trips to cultural events (sponsored by community partners) and extracurricular activities such as the student council, basketball, and a marching drum unit.

The third protective process is caring and support. As noted in the discussion of the first strategy, community partners provide programmatic and financial support. For example, the law firm of Pepper, Hamilton, & Scheetz trains approximately 30 students per year in peer conflict resolution. Also, the school has a Parent Outreach Program to increase parental support for the children's education.

Research shows that specific activities Stanton is using to enhance protective factors can be effective. Foorman et al. (1998) presented evidence of effectiveness for the reading instruction approach that Stanton is using in kindergarten and grade one, but Mac Iver et al. (1998) found in a two-year evaluation that the Sylvan Learning Systems program yielded mixed results. In the Baltimore City Public Schools, students' reading and math scores increased, but only the math scores increased significantly. Nettles (1991a, 1991b) examined findings from evaluations of community-related programs and found that such programs can foster resilience through the provision of social support, resources, and instructional activities, such as tutoring. Hawkins et al. (1992) have suggested that the strategy of enhancing protective factors may be an effective one for alleviating problem behaviors, including substance use, risky sexual behavior, and delinquency. Although the alleviation of such behaviors is not a direct goal, Stanton expects that the enhancement of resilience in the early grades will contribute to the alleviation of problem behavior.

Assessing Paths to Student Resilience

The second strategy, assessment of factors associated with individual resilience, is part of data collection in three Washington, DC, schools in CRESPAR's study of the influence of perception of violence and other stress on school-related outcomes. Student participants were enrolled in third and fourth grades in 1996-97 and attended schools in the southeast quadrant of the city.

CRESPAR researchers began the study in 1995 with an extensive literature review that resulted in the development of a conceptual framework of individual resilience. As shown in Figure 1, this framework included four components: (1) the risk/protective environment, (2) meaning, (3) student investments, and (4) outcomes. The risk/protective environment includes perceptions of interactions and resources within specific contexts, such as the neighborhood, the family, and the peer group, or within a combination of contexts. In turn, perceptions of the environment as risk-laden or protective shape individual meaning, defined as self-related beliefs and reasons for engaging or not engaging in academic tasks. Individual meaning influences options for action (student investments), such as effort at classwork and attending school. The results of these choices can be adaptive (resilient) or maladaptive outcomes.

The conceptualization described above was influenced by the theoretical models of community effects on student attainment (Nettles, 1991b), adolescent risk-taking behavior (Levitt et al., 1991), and educational risk and resilience (Connell et al.,1994). The models differ primarily in the choice of specific variables and in the way in which the action component is defined.

The framework in Figure 1 guided the selection of measures. Measures of the perceived risk/protective environment included The Social Support Scale-Revised (Dubow & Ullman, 1989), which assesses the child's appraisals of peer, family, and teacher support; Life Events and Circumstances (Pryor-Brown & Cowen, 1989), an instrument which assesses stressful events that have occurred in the child's life in the past year (for example, the loss of job by a parent); and the Perceptions of Environmental Violence Scale (Hill, 1991), a set of items that measures the child's perception of violence in the home, the school, and the neighborhood. Measures of meaning included the Self-Description Questionnaire (Marsh, 1990), an instrument that assesses seven dimensions of self-concept (including reading and mathematics) and the Motivations for Reading Questionnaire (Wigfield, Guthrie, & McGough, 1996), which taps different aspects of children's reasons for reading (e.g., reading efficacy, importance of reading). Student investments will be measured by teacher ratings of classroom behavior and resilient outcomes by scores on the ninth edition of the Stanford Achievement Test Series (Stanford 9), the standardized instrument used in the District of Columbia Public Schools. The source of data for these measures will be school archival records. A complete set of measures is available on students enrolled at Stanton during 1996-97 in grades three and four. These data will be used in the CRESPAR study on perceptions of violence and school outcomes.

At the school's request, measures of hypothesized resiliency (protective) factors, such as the Self-Description Questionnaire, were administered in classroom to Stanton students in grades two through six. Data collection procedures and descriptive analyses are described in Appendix A. As discussed next, those data were used to increase staff awareness of the strengths children bring to and develop in the school context. Providing such data is congruent with the strategy of mobilizing protective processes, specifically the process of establishing positive expectations for students' success.


Figure 1.
A Framework for Examining Environment and Individual Factors that Influence Resilient Outcomes

Figure 1.  A Framework for Examining Environment and Individual Factors that Influence Resilient Outcomes

Increasing Awareness and Understanding 

The first event of the Stanton/CRESPAR partnership was the one-day staff development retreat in August 1996. CRESPAR staff designed the retreat, which featured topics requested by the principal and faculty, to introduce and illustrate the constructs of risk, resilience, and protection. In one of three sessions, for example, participants used Benard's (1991) threefold categorization of protective factors (high expectations, opportunities for participation, and caring and support) to generate concrete examples of such factors in the Stanton school and community and in children's families. In a second session, participants discussed the family as a source of protection and of risk, and in the third session, participants discussed the risks that substance use presented for students and families.

Two professional development workshops, in September 1997 and in April 1998, presented opportunities to build on the understandings of the initial workshop, to present the results of research data collected from Stanton students during the 1996-97 academic year, and to discuss Stanton's progress in implementing activities to foster resilience in students. For example, at the September workshop, the handout in Appendix B was distributed to faculty and parents. The handout summarizes how Stanton students described themselves on four protective factors: ability to form positive relationships, love of learning, self-motivation, and self-worth and competence. The handout in Appendix C was distributed to teachers at the April workshop and shows how students in each grade ranked motivations to read. Rankings ranged from 1 (reflecting the highest average on a motivation scale) to 11 (the lowest average on a motivation scale). All grades had the highest average score on the importance of reading, and extrinsic reasons were also rated highly. Students in grades two and six on average reported higher efficacy in reading than students in other grades.

Other efforts to inform the faculty and staff include making reference materials on resilience and other topics available in the school's library, and informal conversations on resilience throughout the school year between CRESPAR researchers and school personnel.

The Stanton Case Study

As the collaborative relationship between Stanton and CRESPAR deepened, we realized that the school's increased actions to provide a protective environment would have to become part of the conceptual framework. As shown in Figure 2, the current framework expands the risk/protective environment to include not only students' perceptions of social support and stress, but the school organization and school-initiated activities as well. The figure may be summarized in the following propositions.

1. Schools can implement activities that mobilize protective processes and mitigate risk factors when the school environment is orderly and safe.A stable level of school organization is one of the prerequisites for successful implementation of innovative programs (Gottfredson & Gottfredson, 1985). We are examining the school's role in fostering student resilience through direct observations in the school, a comparison of plans and accomplishments, and review of documents produced by the school. We will compare results of qualitative analyses of school organization with teacher responses on the Effective School Battery (ESB, Gottfredson, 1991).

The remaining propositions are ones that the CRESPAR research project is exploring in Stanton's students:

2. Students will view the environment as supportive when school resources are increased and activities that enhance protective factors are implemented as designed. 3. Students can view themselves as competent and motivated when they feel supported by parents, the school, and the community. 4. Students' everyday efforts can increase when they view themselves as competent and motivated. 5. Resilience in school settings can occur when students' everyday efforts are increased.

Evidence from research on adolescent populations suggests that Stanton's conceptualization of resilience may be a fruitful one. For example, the educational risk and resilience model (Connell, Spencer, & Aber, 1994) was validated in three separate samples of African American young adolescents living in high-risk environments. The four components of the model are (1) context, (2) self (or meaning-making processes), (3) action, and (4) academic-related outcomes. The investigators found that student perceptions of parental involvement (a contextual variable) influenced self-appraisals (i.e., general self-concept, efficacy in school, and perceived quality of interpersonal relationships). These self processes influenced emotional engagement (e.g., satisfaction with school) and behavioral engagement (e.g., doing homework). Depending on the level of engagement, outcomes were negative (e.g., low grades) or positive (e.g., high attendance).

We are using the single-case design (Yin, 1989) to test the hypothesized paths to resilience shown in Figure 2. Questionnaire responses from Stanton students in grades two through six will be analyzed. To determine effectiveness of the approach in accomplishing schoolwide changes in academic competence, we will use changes in normal curve equivalents (NCEs) on the Stanford 9 reading and mathematics subtests. NCEs are the metric used in reporting achievement gains in Title I programs. Moreover, we will compare changes in proportions of students who met defined performance standards on the Stanford 9. We discuss results on schoolwide impact below.

Figure 2.
The Current Stanton Framework

Figure 2. The Current Stanton Framework

Impact on Student Achievement

As discussed earlier, Stanton targeted district and community resources toward mobilizing protective processes, primarily those pertaining to competence in reading and mathematics. To gauge the school's progress, changes in normal curve equivalents (NCEs) were examined for grades two through six on the Stanford 9. The school staff administered the test in October 1997 (fall) and April 1998 (spring). First graders were given the fourth edition of the Stanford Early School Achievement Test in the fall and the Stanford 9 in the spring. As shown in Table 2, the reading gains from fall to spring in grades two through six were 11 to 21 NCEs. The mathematics gains in these grades ranged from 13 to 30. As shown in Table 2, the reading NCE gain was 4.1 and the mathematics gain was 18.1.

The Stanford 9 report also presents student performance according to four categories or levels defined by expert panels: (1) below basic (little or no mastery of knowledge and skills); (2) basic (partial mastery); (3) proficient (prepared for the next grade); and (4) advanced (superior performance). Schoolwide (grades two through six combined) changes were examined according to two categories of performance standards, (1) below basic and (2) basic, proficient, and advanced. Table 3 shows the frequencies and percentages of students below basic in reading at the fall and spring administrations respectively and students who scored at basic levels and above; Table 4 shows the frequencies and percentages in each category for mathematics.

Changes from fall to spring on both subtests were significant. On the reading subtest, 57% of students taking the test in the fall were below basic, and of students taking the test in the spring, 41.4% were below basic, X2 (1, N = 589) = 14.14, p < .05. In mathematics, 73% of students taking the test in the fall were below basic level, and in the spring 39% were below basic, X2 (1, N = 607) = 68.07, p < .05.

Table 2
Mean National Normal Curve Equivalents on Reading and Mathematics Subtests of the Stanford 9

 

Subtest

 

Reading

Mathematics

Change

Fall 1997

Spring 1998

Change

Fall 1997

Spring 1998

Change

1

43.9*

48.0

4.1

32.5*

50.6

18.1

2

25.4

37.4

12.0

16.6

47.0

30.4

3

21.6

43.4

21.8

19.9

50.5

30.6

4

24.0

35.3

11.3

27.0

41.1

14.1

5

19.2

34.8

15.6

17.9

35.1

17.2

6

29.0

41.3

12.3

25.7

42.9

17.2

Table 3
Performance Levels on the Reading Subtest of the Stanford 9

 

Standard

 

Below Basic

Basic & Above

Test Date

n

%

n

%

Fall 1997

152

57

116

43

         

Spring 1998

133

41

188

59

Note: Percentages are calculated across the rows.

Table 4

Performance Levels on the Mathematics Subtest of the Stanford 9

 

Standard

 

Below Basic

Basic & Above

Test Date

n

%

n

%

Fall 1997

208

73

78

27

         

Spring 1998

125

39

196

61

Note: Percentages are calculated across the rows.

Conclusion

The resilience approach at Stanton started as a straightforward research study, wherein the researchers would come to the school, collect the data, and report the results to the faculty. Instead, resilience is emerging as an organizing principle that gives Stanton a means of integrating school improvement, regular and enhanced curricular offerings, and processes that emphasize caring, high expectation, and opportunity. Moreover, resilience orients our thinking toward an optimistic view of our students' prospects. Elementary school children do not describe themselves as being at risk: someone assesses certain factors and applies the label. Indeed, Stanton sees the problem and deficit-focused language of risk as fundamentally at odds with the goals of school improvement. But until schools adapt a different paradigm-and to us the resilience approach is a new paradigm-we will continue to see children in urban schools as problems in the present and the future.

Although the resilience emphasis has emerged over two and a half years at Stanton, other schools might take an abbreviated route. The essential ingredients are a willingness to examine a new way of thinking, an organizational readiness to fill in the gaps in protective processes through use of effective instructional programs and involvement of parent and community partners, and a way of assessing student factors related to resilience. A teacher checklist for recording information on individual student's resiliency factors is described in Sagor (1996), and Henderson and Milstein (1996) suggest ways of developing profiles of student, faculty, and organizational resilience. We are attempting to identify reliable measures of student protective factors, and future plans call for providing teachers with information on ways to administer select measures, to interpret the data on individual students, and to use data to implement strategies to foster student resilience. We are also examining ways to share the information with families, who too often get dismal news about children's progress.

We have outlined the ways in which one school, Stanton Elementary, is embracing the resilience approach. We do not view this as another program to improve test scores and grades, although the results thus far are promising. Rather, we see this as a way of giving meaning to the phrase, "building on children's strengths." We do this first by assessing the child's view of the environment as caring or threatening and then by identifying protective factors within the child. Stanton is beginning to use this information to understand the inner strengths that chil­dren bring with them to school, to create a more accurate picture of the stress that the school can alleviate in their lives, and to fine-tune activities that will make Stanton a more caring, protective school environment for all students.

References

Anderson, L. (1994). Effectiveness and efficiency in inner-city public schools: Charting school resilience. In M. C. Wang & E. W. Gordon (Eds.), Educational resilience in inner-city America: Challenges and prospects (pp. 141-149). Hillsdale, NJ: Lawrence Erlbaum.

Benard, B. (1991). Fostering resiliency in kids: Protective factors in the family, school, and community. Portland, OR: Western Center for Drug-Free Schools and Communities.

Comer, J. (1985). The Yale-New Haven primary prevention project: A follow-up study. Journal of the American Academy of Child Psychiatry, 23(2), 154-160.

Connell, J. P., Spencer, M. B., & Aber, J. L. (1994). Educational risk and resilience in African-American youth: Context, self, action, and outcomes in school. Child Development, 65, 494-506.

Dubow, E. F., & Ullman, D. G. (1989). Assessing social support in elementary school children: The survey of children's social support. Journal of Child Clinical Psychology, 18, 52-64.

Foorman, B. R., Francis, D. J., Fletcher, J. M., Schatschneider, C., & Mehta, P. (1998). The role of instruction in learning to read: Preventing reading failure in at-risk children. Journal of Educational Psychology, 90(1), 37-55.

Gottfredson, G. D. (1991). The effective school battery: User's manual. Odessa, FL: Psychological Assessment Resources.

Gottfredson, G. D., & Gottfredson, D. C. (1985). Victimization in schools. New York: Plenum.

Hawkins, J. D., Catalano, R. F., & Miller, J. Y. (1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychological Bulletin, 112, 64-105.

Henderson, N., & Milstein, M. M. (1996). Resiliency in schools: Making it happen for students and educators. Thousand Oaks, CA: Corwin.

Hill, H. (1991). Perceptions of environmental violence scale. Washington, DC: Howard University.

Levitt, M. Z., Selman, R. L., & Richmond, J. B. (1991). The psychosocial foundations of early adolescents' high risk behavior: Implications for research and practice. Journal of Research on Adolescence,1(4), 349-378.

Mac Iver, M. A., Stringfield, S., McHugh, B., & Snively, F. (1998). Implementation and effects of the Sylvan Learning Systems Program in the Baltimore City Public Schools: Evaluation report. Baltimore: Johns Hopkins University, Center for Social Organization of Schools.

Marsh, H. W. (1990). Self description questionnaire-I: SDQ manual. Campbelltown, Australia: University of Western Sydney, Macarthur.

Masten, A. (1994). Resilience in individual development: Successful adaptation despite risk and adversity. In M. C. Wang & E. W. Gordon (Eds.), Educational resilience in inner-city America: Challenges and prospects (pp. 141-149). Hillsdale, NJ: Lawrence Erlbaum.

Masten, A., Best, K. M., & Garmezy, N. (1990). Resilience and development: Contributions from the study of children who overcame adversity. Development and Psychopathology, 2, 425-444.

Nettles, S. M. (1991a). Community contributions to school outcomes of African-American students. Education and Urban Society, 24(1), 132-147.

Nettles, S. M. (1991b). Community involvement and disadvantaged students: A review. Review of Educational Research, 61, 379-406.

Nettles, S. M., & Pleck, J. H. (1994). Risk, resilience, and development: The multiple ecologies of black adolescents in the United States. In R. J. Haggerty, L. Sherrod, N. Garmezy, & M. Rutter. (Eds.), Stress, risk, and resilience in children and adolescents: Processes, mechanisms, and intervention (pp. 147-181). New York: Cambridge University Press.

North Central Regional Educational Laboratory. (1994). Resilience research: How can it help city schools? CITYSCHOOLS, I(1), 11-18.

Pryor-Brown, L., & Cowen, E. L. (1989). Stressful life events, support, and children's school adjustment. Journal of Clinical Child Psychology, 18, 214-230.

Rutter, M. (1987). Psychosocial resilience and protective mechanisms. American Journal of Orthopsychiatry, 57, 316-331.

Sagor, R. (1996). Building resiliency in students. Educational Leadership, 54(1), 38-43.

Wang, M. C., & Gordon, E. W. (Eds.). (1994). Educational resilience in inner-city America: Challenges and prospects. Hillsdale, NJ: Lawrence Erlbaum.

Wang, M. C., Haertel, G. D., & Walberg, H. J. (1994). Educational resilience in inner cities. In M. C. Wang & E. W. Gordon (Eds.), Educational resilience in inner-city America: Challenges and prospects (pp. 41-149). Hillsdale, NJ: Lawrence Erlbaum.

Werner, E. E. (1990). Protective factors and individual resilience. In S. Meisels and J. Shonkoff (Eds.), Handbook of early childhood intervention. New York: Cambridge University Press.

Wigfield, A., Guthrie, J., & McGough, K. (1996). A questionnaire measure of reading motivations. (Instructional Resource No. 22). Athens, GA: National Reading Research Center.

Winfield, L. F. (Ed.). (1991). Resilience, schooling, and development in African-American youth: A conceptual framework. Education and Urban Society, 24(1), 5-14.

Yin, R. K. (1989). Case study research: Design and methods. Newbury Park: CA: Sage Publications.

Zimmerman, M. A., & Arunkumar, R. (1994). Resiliency research: Implications for schools and policy. Social Policy Report: Society for Research in Child Development, 8(4).

Appendix A

Assessing Risk and Protective Factors

This appendix presents descriptive data on four questionnaires administered to groups of students at Stanton Elementary School in Washington, DC. These questionnaires include three measures to assess protective (resiliency) factors, including the Social Support Appraisal scale, the Self-Description Questionnaire, and the Motivations for Reading Questionnaire. In addition, a measure of stress, the Life Events and Circumstances scale, was administered.

Method

Participants

Participants were 263 children in the second through sixth grades. Parents were notified of the assessment according to guidelines in the District of Columbia Public Schools memorandum, Collaborative Partnership with the Center for Research on the Education of Students Placed At Risk, December 8, 1995.

Procedure

Students completed paper-and-pencil measures in their classrooms monitored by four research assistants (3 female and 1 male). All the measures were read aloud to minimize any problems related to reading. The research assistants worked in pairs. While one research assistant read aloud, the other assistant walked around the classroom checking to make sure that the students were following directions, not skipping ahead, or checking more than one response. The teachers were asked to leave the classroom during the questionnaire session to protect the students' confidentiality.

Data were collected in October and November of 1996 and in January of 1997.

Measures

The Social Support Appraisal Scale-Revised (Dubow & Ullman, 1989) is a 41-item pencil-and-paper instrument that assesses the child's appraisals of peer, family, and teacher support. Items were developed to reflect an individual's conceptualization of social support-information indicating to the individual that he or she is valued and esteemed by others. Sample items illustrating the content of the three major subscales include: peer items (e.g., whether the child feels left out by his/her friends), family items (e.g., whether the child is an important member of his/her family), and teacher items (e.g., whether the child feels his/her teachers are good to ask for advice or help about problems). The following sample item illustrates the format of each item: "Some kids feel left out by their friends, but other kids don't. Do you feel left out by your friends?" The child responds to each item on a 5-point continuum (1 = always, 5 = never). Dubow and Ullman (1989) reported an internal consistency reliability (Cronbach's alpha) of .88 and 3-4 week test-retest reliability of .75.

The Self-Description Questionnaire-I (SDQ-I, Marsh, 1990) is designed to measure seven dimensions of self-concept. The instrument consists of 72 short items to which students respond along a 5-point response scale where 1 = "false," 5 = "true." The following is a brief description of each of the seven subscales:

1. Physical abilities/sports, on which students rate their ability and enjoyment of physical activities, sports, and games.

2. Physical appearance, wherein students rate their own attractiveness, how their appearance compares with others, and how others think they look.

3. Relationship with peers, wherein students rate how easily they make friends, their popularity, and whether others want them as a friend.

4. Relationship with parents, wherein students rate how well they get along with their parents, whether parents are easy to talk to, like them, and whether they like their parents.

5. Reading, on which students rate their ability and their enjoyment/interest in reading.

6. Mathematics, wherein students rate their ability and enjoyment/interest in mathematics.

7. General school subjects, on which students rate their ability and enjoyment/interest in "all school subjects."

The SDQ-I includes three composite scores: (1) Total Nonacademic Self-Concept, which is the mean of responses to the physical abilities, physical appearance, relations with parents, and relations with peers scales; (2) Total Academic Self-Concept, which is the mean of responses to the reading, mathematics, and general school scales; and (3) the Total Self-Concept, the mean of the responses to the seven factors named above.

The Motivations for Reading Questionnaire (MRQ, Wigfield, Guthrie, & McGough, 1996) is a 54-item questionnaire that assesses different aspects of children's motivation for reading. MRQ consists of 11 subscales, with each item a short sentence to which the child responds on a 4-point scale (1 = very different from me, 4 = a lot like me). The 11 subscales are as follows: Reading efficacy (e.g., I am a good reader), Reading challenge (e.g., I usually learn difficult things by reading), Reading curiosity (e.g., I like to read about new things), Aesthetic enjoyment of reading (e.g., I read stories about fantasy and make believe), Importance of reading (e.g., it is very important to me to be a good reader), Compliance (e.g., I read because I have to), Reading recognition (e.g., I like having the teacher say I read well), Reading for grades (e.g., I read to improve my grades), Social reasons for reading (e.g., I often read to my brother or my sister), Reading competition (e.g., I like being the best at reading), and Reading work avoidance (e.g., I don't like vocabulary questions).

Internal consistency reliabilities of these scales were mostly greater than .70 save for reading efficacy, importance of reading, reading recognition, compliance, and reading work avoidance (Wigfield, Guthrie, & McGough, 1996).

The Life Events and Circumstances Scale (Pryor-Brown & Cowen, 1989) is a 22-item instrument that assesses stressful life events that have occurred in the child's life within the past year (e.g., child changed schools, best friend moved out of town, loss of job by parent). The following sample items illustrate the format of each item: "I had to change to a new school," "My best friend moved out of town." The child responds in a dichotomous way by checking either "yes" or "no." Nineteen of the items represent events over which the child has very little or no control (e.g., parents separated, parent lost a job). These items are thus less likely to be confounded with the resiliency outcomes as compared to the three events such as "a bad mark on a test."

Results

Descriptive Statistics

Table A1 presents the means, standard deviations, and ranges of each of the measures. Students in the combined sample reported an average of 7.03 stressful life events. The second, third, and fourth grades averaged 7.63, 7.48, and 7.14 life events, respectively. Fifth and sixth graders reported lower mean life events (6.04 and 6.95, respectively). Second grade students reported the highest mean level of perceived social support (170.3), and fourth graders the lowest (159.9). The mean score for the combined sample was 163.49.

Scores on the Self-Description Questionnaire for the combined sample were 36.09, 36.02, and 35.79 for the Total Nonacademic, Total Academic, and Total Self scores, respectively. These means were higher than those (31.77 for Total Nonacademic, 29.53 for Total Academic, and 30.89 for Total Self) reported for the normative sample of 3,562 children in New South Wales, Australia (Marsh, 1990).

Table A1
Means, Standard Deviations, and Ranges of Study Variables for the Combined Samples and for Each Grade

 

Grades

Variables

Combined Sample

2

3

4

5

6

Perceived Social Support

M

163.49

170.3

159.9

158.2

162.3

166.1

 

SD

24.68

24.6

23.4

24.2

24.9

25.5

 

Range

90-205

90-203

100-193

115-201

101-205

101-205

 

N

252

68

58

46

46

34

Life Events

M

7.03

7.63

7.48

7.14

6.04

6.95

 

SD

3.69

3.32

3.24

4.47

3.83

4.31

 

Range

0-22

1-15

1-14

0-23

0-19

0-22

 

N

259

68

56

50

48

37

Total Nonacademic

M

36.09

38.63

35.90

34.90

35.2

34.4

 

SD

4.09

1.89

4.38

4.00

4.34

4.43

 

Range

15-40

32-40

15-40

23-40

21-40

24-40

 

N

233

60

53

45

42

33

Total Academic

M

35.07

38.63

35

33.4

33.3

33.3

 

SD

5.28

2.74

5.01

1.90

4.19

5.84

 

Range

12-40

21-40

15-40

23-39

12-40

23-40

 

N

233

60

53

45

42

33

Total Self

M

35.79

38.88

35.7

34.2

34.6

34.1

 

SD

4.41

2.09

4.47

4.05

4.91

4.54

 

Range

15-40

27-40

15-40

23-40

17-40

24-40

 

N

233

60

53

45

42

32

Motivations for Reading

M

34.68

36.96

35.54

34.40

33.85

31.34

 

SD

4.63

2.98

4.21

3.68

4.48

4.98

 

Range

9-40

28-40

19-40

23-39

18-40

20-39

 

N

256

69

56

47

47

35

* p < 0.10  ** p > 0.05  *** p < 0.01

Mean total scores on the Motivations for Reading questionnaire were 34.69 for the combined sample (grades 2 through 6); 36.96, 35.54, and 34.4 for the second, third, and fourth grade cohorts, respectively, and 33.86 and 31.34 for the fifth and sixth grades, respectively.

Correlations among Study Variables

Table A2 presents the intercorrelations among study variables for all the students in the sample. Perceived social support was significantly correlated with motivations for reading and academic self-concept. However, social support was not related to total nonacademic and total self scores. Social support and life events were negatively correlated; students in the combined sample who reported a high level of stressors also perceived that social support was low. Life events were unrelated to each of the self-related beliefs and attitudes.

The table shows high correlations, as expected, between the components of the Self Description Questionnaire (total nonacademic, total academic, and total self). These self conceptions were also positively related to motivations for reading scores, but the associations were modest.

Table A2
Intercorrelations, Significance Levels, and Sample Size of All Study Variables
(Grades 2-6)

 

Grades

Variables

1

2

3

4

5

6

1. Perceived Social Support

 

--

0.24***

0.07

0.14**

0.10

0.18***

 

r

252

 

222

222

2229

 
 

N

 

249

     

248

2. Life Events

   

--

0.07

0.10

0.09

0.09

 

r

 

259

231

231

231

254

 

N

           

3. Total Nonacademic

r

   

--

0.71***

0.89***

0.15**

           

233

227

 

N

     

233

   

4. Total Academic

       

--

0.92***

0.16***

 

r

       

233

 
 

N

         

227

5. Total Self

r

       

--

 
           

233

0.17***

 

N

         

227

6. Motivations for Reading

           

--

 

r

         

256

 

N

           

* p < 0.10  ** p > 0.05  *** p < 0.01

Appendix B

Handout Distributed at Professional Development Retreat, September 1997

PROTECTIVE FACTORS IN STANTON STUDENTS

  • Ability to Form Positive Relationships
    Of the many ways that Stanton students describe themselves, in all grades except the 3rd grade, Parent Relations ranked highest. Children thought they got along well with parents, liked them, and experienced parental acceptance and approval.
  • Love of Learning
    Ability, enjoyment, and interest in reading (3rd and 4th graders) and math (2nd and 6th graders) ranked high (3rd of 6) on the list of dimensions of self-concept.
  • Self-Motivation
    The number one motivation for reading among Stanton students is the importance of reading. Their desire for outperforming others ranked second among the motivations.
  • Self-Worth and Competence
    On the dimensions of self-concept, physical attractiveness ranked second. Other dimensions included physical abilities, peer relations, parent relations, reading, and mathematics.

Overall, Stanton students perceived themselves as capable and proud of the way they were.

Appendix C

Handout Distributed to Stanton Faculty

Table 1
How Stanton Students Rate Motivations for Reading

Motivation Scale

Sample Item

Grade 2

Grade 3

Grade 4

Grade 5

Grade 6

Importance of Reading

"It is very important to me to be a good reader."

1

1

1

1

1

Reading Competition

"I like being best at reading."

2

2

2

4

3

Reading for Grades

"I read to improve my grades."

4

3

3

2

2

Reading for Recognition

"I like having the teacher say I read well."

5

4

4

6

3

Reading Efficacy

"I am a good reader."

3

7

5

7

3

Reading Curiosity

"I like to read about new things."

5

6

6

3

6

Reading Involvement

"I make pictures in my mind when I read."

5

4

7

5

5

Reading Challenge

"If a book is interesting, I don't care how hard it is to read."

7

8

7

8

4

Social Reasons for Reading

"I often read to my brother or my sister."

6

5

7

10

8

Compliance

"I read because I have to."

8

9

8

9

7

Reading Work Avoidance

"I don't like vocabulary questions"

9

10

9

11

9